I immediately thought there must be some clever combination of query modifiers that could roughly reproduce the discussion filter behavior... Of course I'm not the first one with this idea— there seems to be browser extensions[1][2] that accomplish this by appending your search query with:
intext:forum "post"|inurl:forum|"posts:"|inurl:viewtopic
[1]: https://chrome.google.com/webstore/detail/discussions-button...
[2]: https://addons.mozilla.org/en-US/firefox/addon/google-forum-...
Could you provide more detail on this search filter? I haven't heard about its removal before and it sounds interesting.
https://productforums.google.com/forum/#!msg/websearch/Psb6O...
https://www.reddit.com/r/google/comments/2b54ux/google_compl...
http://www.realitymod.com/forum/f11-off-topic-discussion/130...
...etc.
https://productforums.google.com/forum/#!topic/webmasters/0Y...
That's probably ... correct.
None of it seems to be conspiracy theory or low-quality material.
I'd classify that as conspiracy theory.
Even if it were, if it "wins" the pagerank contest it should be represented in the results accordingly.
I don't go to google for a human-curated yahoo-style directory service. I go to google for search - specifically for pagerank-style search results.
https://www.amazon.com/Russian-Revolution-New-History-ebook/...
The Russian revolution was co-opted by Lenin and the Bolsheviks who were funded by the German foreign office and more than anything else were unscrupulous about who they murdered. 25 million people were dead by the end of the revolution (early 20s).
There's not so much as a hint of that just scanning though the titles.
> These findings make clear that the decline in Google search traffic to the WSWS is not the result of some technical issue, but a deliberate policy of censorship.
...or that they have not been keeping up with their SEO, are getting edged out by AMP sites, tripped over some other google rule, etc...
The change has effected a broad array of left-wing websites, with different SEO methods. It has also happened directly after Google's new algorithm, which stated clearly what they were planning on doing.
The irony: they prefer to cry wolf instead of doing SEO and don't want to be called a conspiracy theory site.
"Socialist news" would be one of the searches that "indicates the user is seeking an alternative viewpoint."
Quite a claim.
Obviously it would be great if others could compete, but I think it's simply a market gap (innovation, capital investment, etc) at this point and not for lack of competitors trying. Many died out (Yahoo!, Excite, Altavista), some survive as niche (DuckDuckGo), some still exist (Bing, Ask/Jeebs), and some are still massive but serve other markets (Baidu, Yandex).
Google has become so successful that their search is very advanced compared to upstarts. The more Google learns from their massive index corpus and userbase, the further ahead they get. Baidu might have a comparable dataset and userbase, but they specialize in different markets with different languages and cultures.
Not necessarily (distributed economic planning, comumnal planning and indeed some (though not I) advance market Socialism), and further, things like search can be decentralised, as I'm sure mosts Socialists would agree with if aware that such technology exists.
>If purist socialists had there way, wouldn't there only be one search engine?
No, I see no reason to think that. This is largely stemming from the myth that "everyone has one car under Socialism"; it neglects that aside from seeking profit, different cars have different uses and advantages. It's probably the same with search engines and GNU/Linux distros.
>Google has become so successful that their search is very advanced compared to upstarts.
I agree, and I think that in a world where there is less incentive to keep secrets (though of course there is always incentive, such as personal pride, mastery, or even embarrassment over how your code looks!) because profit doesn't need to be protected from "theives", such research might even be public for others to benefit from. The learning technology can be free. Google would not have to collect data "on the sly", perhaps.
However the response to this is - would people willingly (and by this I mean when properly and fully informed of what they are doing) give over this data that Google uses to learn in such a way? If the answer is no, perhaps it's time to consider the ethical ramifications of such 'learning'.
If you would like to know more about the project, the blog is here: http://blog.levelnews.org
Much of the codebase is open source and can be found on GitHub if you'd like to contribute: https://github.com/levelnewsorg
Regulation is done on utilities like cell providers, cable/internet providers and consumers are usually aware of multiple choices. If I ask a non-tech user to name a search engine other than Google, they likely can't. So users do their search and interact with the results they are presented.
Edit: Removed my personal experience. Added thoughts on regulation.
Even if we get to the point that certain pages are only likely to be found via 10-20 year old technologies, so what? Those technologies were invented for text articles and still work well for them, and text articles are what matter most. Similar things are also true about the runners-up in "matter most", which in some order are straightforward image, sound and video files.
The word is becoming meaningless dribble in the hands of axe-grinding cretins who believe that everyone must be forced to read their material.
But it was probably naive to assume Google would limit itself to censoring political opinions with which one disagrees.
Most people on the actual left weren't caught up in the "fake news/Russia/Russian fake news stole the election" hysteria because none of them supported Clinton in the first place.
As the article said:
> In a set of guidelines issued to Google evaluators in March, elaborated in April by Google VP of Engineering Ben Gomes, the company instructed its search evaluators to flag pages returning “conspiracy theories” or “upsetting” content unless “the query clearly indicates the user is seeking an alternative viewpoint.” The changes to the search rankings of WSWS content are consistent with such a mechanism.
Notice how Google said "alternative viewpoint". Alternative to what? Google seems to be establishing a new orthodoxy.
You can think of it this way, do you really really want some particular corporation or government agency to have that tool in their toolbox? Because if they do they'll use it.
Kinda like right to work laws designed to prevent 'illegal' emigrants from working. You really really want the government to be able to prevent you from getting a job?
Salon/AlterNet: Fighting fake news: Google’s new fact-check tool attempts to combat a global issue
http://www.salon.com/2017/04/23/fighting-fake-news-googles-n...
Salon/AlterNet's main concern was apparently that it didn't do enough.
> this doesn’t do anything to directly combat the use of Google’s platform to spread false or offensive stories.”
When a source repeatedly tells you not to believe your own eyes, but offers no valid counter-argument--you should be suspicious. I'm fine with Google/Facebook or whoever filtering out known dishonest and inaccurate websites, whether they are intentionally malicious or just ignorant.
Your experience is irrelevant, because we're talking about a particular case, not a hypothetical website. Do you have evidence that your experience applies in this particular case, or are you just generalising? For a user named after a philosopher, you don't seem to be very keen to apply logic in this circumstance, perhaps because of bias or bad prior experience.
Not to mention you're grouping in several tendencies and ideologies and opinions within a very broad movement that's lasted 200 years to date in your criticism, which is at best inaccurate and at worst disingenuous.
It's also true in this case. WSWS is a political and a ideological advocacy site with a non-mainstream interpretation of world history. If you want an objective view of the Russian Revolution, this is not the site you would go to - so I have no issues with Google tuning their search engine to suppress them in the search results for those kinds of queries.
Given the amount of power and money they now have, effectively having "won the game," it seems ethical to me to open-source their entire search algorithm.
They would have to open source the indexed data and training data too, assuming you wanted to know why site A doesn't rank well for term X.
My original comment remains unedited below.
--
For a concrete demonstration of pathological de-ranking, do a query for "site:web.archive.org".
I get "59,000 results" on page 1, but page 2 will never load!
There are a few results, which proves that a) web.archive.org are not using robots.txt or other blocking techniques, and b) that Google's infrastructure is inhaling content. But it's invisible.
Think about how sad this is - once a site goes dead, it's offline, even though the content is still publicly accessible. If only that context was indexed using a decent search engine.
Practically speaking, I totally acknowledge that archived content is complex to surface; sites can be pulled offline because content needs to be disappeared for any number of reasons, etc. I recognize the general difficulty of getting this right. So I'm not _really_ arguing "if only this were surfaced", because it's unfair to - I'm more saying "hey look, this is what it looks like when something has been completely killed," as a demonstrable and extreme datapoint.
That search works just fine for me.